銀河リッジ x 線放射の中性鉄輝線における低エネ...

銀河リッジ X 線放射の中性鉄輝線における低エネルギー宇宙線の寄与

信川久実子 (奈良女子大)

信川正順 (奈良教育大)、山内茂雄 (奈良女子大)、内山秀樹 (静岡大)、小山勝二 (京大)

© ISAS/JAXA

すざく

天の川銀河に拡がるX線放射 2

銀河中心・バルジ・リッジに個々の天体に分解できない拡がったX線放射が存在

銀河中心・銀河面からの中性鉄輝線（6.4 keV）3

銀河中心 (|l|<1°) 銀河面

● 銀河中心 (GCXE) の 6.4 keV輝線は分子雲と相関、6.4 keV輝線強度が時間変化 ● Sgr A*からの過去のX線フレアを分子雲が反射「X線反射星雲」 (Koyama+96, Murakami+00, M. Nobukawa+11)

高温プラズマ起源 kT~ 数千万-1億 K

銀河面上にも6.4 keV放射 ̶> 起源は未解明

(Uchiyama+13)

6 7

He状鉄 @6.7 keVH状鉄

@7.0 keV

中性鉄@6.4 keV

低温ガスから放射背後に高エネルギー現象

(Uchiyama+13)

点源に分解？ 4

チャンドラ衛星の (l, b) = (0.0°, –1.4°) の観測 => 鉄輝線の70--80%を点源に分解（Revnivtsev+09、Hong12）

Intensity

10‒4

Resolved fraction

1

0.8

0.6

0.4

0.2

0

Energy (keV)4 6 8

全体

分解した点源

未分解

4 6エネルギー (keV)

Intens

ity

10-4

(Revnivtsev+09)

● 強磁場・弱磁場激変星、コロナ星の連星系の集まり? ● 6.4 keVの起源 = 点源？--> 3本の鉄輝線を分離していない ● 非常に狭い範囲の観測 ̶> 空間構造がカギ

バルジ

(Revnivtsev+09)

with statistical significance .4s (minimum number of counts persource is about 10). In Fig. 2a we show the energy spectrum of thetotal emission from HRES, as well as the two components associatedwith the detected sources and with the remaining unresolved emis-sion. Most importantly, the summed spectrum of detected sourcesexhibits a pronounced ,6.7 keV iron emission line, a distinctivefeature of the GRXE which was often regarded as an important argu-ment in favour of it being the emission of a truly diffuse hotplasma3,11. But now we clearly see that the bulk of the 6.7 keV lineemission, as well as of the neighbouring continuum, is in fact pro-duced by point sources. We note that apart from the dominant6.7 keV line, the unresolved (partially due to finite energy resolutionof the instrument and due to limited statistics of the observation)blend of lines at 6–7 keV may contain some contribution from6.4 keV iron fluorescent emission, part of which may be unrelatedto the GRXE and result from irradiation of the interstellar medium bydiscrete X-ray sources23,24.

The derived fraction of the X-ray emission resolved into pointsources is shown as a function of energy in Fig. 2b. In the narrow

energy band 6.5–7.1 keV that contains the iron emission line,84 6 12% of the total X-ray emission is resolved. Moreover, we recallthat the remaining unresolved X-ray emission contains a non-negligible contribution from the CXB. Assuming that the intensityof this unresolved component in our 1 Ms Chandra observation is thesame as in the Chandra extragalactic deep fields19 (ICXB, unres 1 Ms 5(3.4 6 1.7) 3 10212 erg s cm22 deg22 in the 2–8 keV energy band, or(2.9 6 1.4) 3 10213 erg s cm22 deg22 at 6.5–7.1 keV, assuming apower-law spectral shape with C 5 1.4), we can estimate that4 6 2% of the total intensity in the 6.5–7.1 keV band is unresolvedCXB emission. We conclude that we have resolved as much as88 6 12% of the GRXE emission into point sources at energies nearthe 6.7 keV line, the feature that was previously used as the strongestargument in favour of a diffuse origin for the GRXE.

Apart from a small contribution from extragalactic sources (about40–50 sources out of 473), most of the sources detected by Chandrain HRES are probably accreting white dwarfs (with luminositiesL2210 keV < 1031–1032 erg s21) and binary stars with strong coronalactivity (with L2210 keV , 1031 erg s21). Indeed, if we plot the fractionof the total GRXE flux contained in sources with fluxes higher than avariable detection threshold (see Fig. 3), the resulting dependenceproves to be in good agreement with the expectation based on theluminosity function of faint X-ray sources measured in the Solar

17:51:20.017:51:30.017:51:40.0

–29:

37:0

0.0

–29:

36:0

0.0

–29:

34:0

0.0

–29:

33:0

0.0

–29:

32:0

0.0

1 arcmin

17:51:20.017:51:30.017:51:40.0

–29:

37:0

0.0

–29:

36:0

0.0

–29:

34:0

0.0

–29:

33:0

0.0

–29:

32:0

0.0

1 arcmin

Figure 1 | The Chandra image in the 0.5–7 keV energy band. Circles ofradius 2 arcsec denote the positions of point sources detected after 1 Msexposure. The Chandra data were reduced following a standard procedure25.The detector background was modelled using the stowed data set (http://cxc.harvard.edu/contrib/maxim/stowed) and adjusted to the conditions ofthe current observations using the count rate at energies 9–12 keV, whereChandra has almost zero effective area. The total measured X-ray surfacebrightness in HRES is I327 keV 5 (4.6 6 0.4) 3 10211 erg s21 cm22 deg22 inthe 3–7 keV band, or equivalently I2210 keV 5 (8.6 6 0.5) 3 10211

erg s21 cm22 deg22 in the 2–10 keV band. Throughout the field there arenoticeable variations of the soft X-ray (,2 keV) surface brightness,due to what appears to be a previously unknown supernova remnantshell projected onto the Chandra field. We note that if a 1 Ms Chandraobservation were repeated in a nearby field, the measured X-ray surfacebrightness would be slightly different because the number of brightest pointsources varies from field to field, an effect known as cosmic variance inextragalactic studies. For the same reason, there may be subtle field-to-fieldvariations in the GRXE spectral shape, and in particular in emission lineratios, and recent observations indicated that such variations do exist10.Additional variations of the spectrum of the unresolved Galactic X-rayemission can be caused by the presence of genuine diffuse X-ray emitterssuch as supernova remnants.

Inte

nsity

10–4

a

b

Res

olve

d fr

actio

n

1

0.8

0.6

0.4

0.2

0

Energy (keV)

4 6 8

Figure 2 | GRXE spectrum and its resolved fraction. a, Spectra collected byChandra within the HRES region. Black data points, error bars and thehistogram show the spectrum of the total emission from HRES; the collectivespectrum of all detected sources is presented in blue and the spectrum of theremaining unresolved emission in the current observations is in red. Theintegrated spectrum of detected sources exhibits a strong ,6.7 keV ironemission line, characteristic of hot (with temperatures 10–1003106 K)plasma emission. This line has been the main support for the popularhypothesis that the GRXE has a truly diffuse, interstellar origin, even thoughsuch hot interstellar plasma cannot be confined within the Galaxy by itsgravitational potential. We took into account that a small fraction ofphotons, X (10% at energies 4–6 keV, according to the ChandraProposers’ Observatory Guide26) from a point source are scattered by thetelescope outside the surrounding circle of radius 2 arcsec. We thereforecorrected the directly measured collective spectrum of detected sourcesF1(E) using the formula ~FF1 Eð Þ~ F1 Eð Þ{F2 Eð ÞA1=A2½ $= 1{X{XA1=A2½ $,where F2(E) is the spectrum of the unresolved X-ray emission, A1 (,2%of the total) is the area covered by the circles of radius 2 arcsec used forcollecting the source fluxes, and A2 is the area outside these circles.b, Fraction of the X-ray emission resolved by Chandra into point sourcesas a function of X-ray photon energy. Error bars in a and b are 68%confidence intervals.

NATURE | Vol 458 | 30 April 2009 LETTERS

1143 Macmillan Publishers Limited. All rights reserved©2009

チャンドラの観測視野

半径～2分角

(Revnivtsev+06)

鉄輝線分布 5

RXTE衛星の観測で、鉄輝線の分布が星の分布と一致 (Revnivtsev+06)

我々の手法: ① 鉄輝線を3本に分離し、 ② それなりに良い角度分解能で ③ 銀河面全体を観測する

銀河面の6.4 keVの起源を探る！

RXTEの角度分解能は1度近赤外マップの角度分解能は0.7度銀河円盤 (星・分子ガス) の厚み (~100 pc = 0.7度)と比較できない

3本の鉄輝線に分離していない

鉄輝線近赤外

（≒ 星の分布）

銀河面

「すざく」でえた鉄輝線の銀経分布 6

10−8

10−7

10−6

10−5

10−8

10−7

10−6

10−5

10 1000

0.51

1.5

Galacitc longitude (degree)1

X−ray

flux (

ph cm

−2 s−1 arcm

in−2)

Ratio

202

6.4 keV

6.7 keV

銀経の絶対値（度）

6.4 keV / 6.7 keV

東西

銀河中心東側銀経-20度付近

Yamauchi, KN+ 2016

6 76 7

10−8

10−7

10−6

(pho

tons

s−1

cm−2

arc

min

)−2X-

ray F

ulx

10−9

−2 −4Longitude (degree)

24 3 −31 −1銀経（度）

対称分布

6.4 keV

6.7 keV

東側の6.4 keVは対称成分 + 非対称成分

銀河中心東側の超過 7

KN+15、ApJL

KN+15、ApJL銀河中心東側の超過 8

−2 −4Longitude (degree)

24 3 −31 −1

CO In

tens

ity (K

km/s)

12

103

104

102

105

10−8

10−7

10−6

(pho

tons

s−1

cm−2

arc

min

)−2X-

ray F

ulx

10−9

対称分布 + 12CO × α

東西

6.4 keV

12CO

6.4 keVの超過成分は分子ガス起源

銀経（度）

対称分布

分子ガスにMeV宇宙線陽子が衝突

Clump 2 CMZ

12CO積分強度図 (Torii+10)東側には濃い分子ガス

銀経（度）

銀緯（度）

銀経-20度付近の超過 9

Churchwell+2009

銀経 -20度

Near 3 kpc arm

12COマップ

Galactic longitude (degree)−20 −22 −24−18

Galac

tic la

titude

(deg

ree)

−1

−2

0

1

2

K km

/s

0

100

200

300

400

すざくの視野

● Near 3 kpc armの接線方向 → 見通す物質量が多い ● 6.4 keV 超過に対応する分子ガスの濃い領域が存在 ● 銀経-20度の超過も分子ガスと相関 ̶> 宇宙線の衝突が最も可能性高い

10−8

10−7

10−6

10−5

10−8

10−7

10−6

10−5

10 1000

0.51

1.5

Galacitc longitude (degree)1

X−ray

flux (

ph cm

−2 s−1 arcm

in−2)

Ratio

202

6.4 keV

6.7 keV

銀経の絶対値（度）

6.4 keV / 6.7 keV

東西

超過以外の6.4 keV輝線の起源は？

「すざく」でえた鉄輝線の銀経分布 10

Yamauchi, KN+ 2016

スケールハイトの測定 11

Yamauchi, KN+ 2016

lb スケールハイト (@ e−1)

スケールハイト

(pc)

0

50

100

150

200

6.4 keV 6.7 keV 白色矮星コロナ星連星系 12CO12

70±17

142±17 130−160

150−300

92±1

6.4 keV: 分子雲に近い6.7 keV: 点源と無矛盾

激変星コロナ星の連星系

(e.g. Patterson+98, Gilmore+00) (Torii, private communication)@ 1σ error

スペクトルの形12

M. Nobukawa, KN+2016

4. DISCUSSION

4.1. Equivalent Widths of the Iron K-shell Lines from mCVs,non-mCVs, ABs, and GDXE

We have determined the mean EW6.40, EW6.68, and EW6.97,with respective values of 169±5 eV, 118±5 eV, and60±4 eV for the mCVs, 82±7 eV, 451±10 eV, and167±9 eV for non-mCVs, and 28±5 eV, 327±8 eV, and45±6 eV for ABs, respectively (Model A, Table 3). TheseEWs are consistent with, but are more accurate with smallererrors than the previous reports. We further detected theNi XXVII-Heα, Fe XXV-Heβ, and Fe XXVI-Lyβlines in themCVs spectrum, and Ni XXVII-Heα and Fe XXV-Heβ lines inthe non-mCVs and ABs spectra for the first time.

We have also obtained high-quality spectra of the GDXE,and accurately determined EW6.40, EW6.68, and EW6.97. Inaddition, we detect many K-shell lines of iron and nickel suchas Ni I-Kα, Ni XXVII-Heα, Fe XXV-Heβ, Fe XXVI-Lyβ,Fe XXV-Heγ and Fe XXVI-Lyγlines from the GCXE spectra.From the GBXE and GRXE spectra, the newly detected linesare Ni XXVII-Heα, Fe XXV-Heγ, and Fe XXVI-Lyγ.

The EW6.40, EW6.68, and EW6.97of the mCVs, non-mCVs,and ABs have been reported by several authors, mainly withASCA, Chandra, XMM-Newton and Suzaku (Yamauchiet al. 2016 and references therein). However, the mean valuesof the EWs have large errors, except for Xu et al. (2016), andhave large variations from author to author and/or frominstrument to instrument. These large systematic errors wouldbe due to different analysis processes from author to author forthe rather faint iron K-shell structures: different instrument,different criteria of the data selections, reductions, the NXBestimations, different analysis tools, and various otherconditions.

We have estimated the EWs using the same instrument (XIS)with unified data reduction and analysis for all the GDXE,mCVs, non-mCVs, and ABs spectra. Therefore, the systematicerrors of EWs, in particular relative systematic errors of EWsamong the GCXE, GRXE, GBXE, mCVs, non-mCVs, andABs would be far smaller than those of the previous reports.This is essential for the reliable reconstruction of the GCXE,GRXE, and GBXE spectra by the combination of the meanspectra of the mCVs, non-mCVs, and ABs.

4.2. Galactic Bulge X-Ray Emission (GBXE)

Using the deep Chandra observations (∼1Ms) at

* * = -l b, 0 .1, 1 .4( ) ( ), Revnivtsev et al. (2009); Hong(2012) made the plot of the integrated point source flux

(6.5–7.1 keV) as a function of the threshold luminosity (2–10keV). They concluded that more than ∼80% flux of the centralregion is resolved into point sources (Figure 3(b) of Hong(2012)). However, a problem is that number fractions(observed XLF) of the CVs and ABs are significantly differentbetween these two authors.The high-quality GBXE and point source (mCVs, non-

mCVs, and ABs) spectra with accurate EW6.40, EW6.68, andEW6.97 obtained in this paper enable us to adopt a differentapproach to the point source origin for the GBXE. The GBXEspectrum, particularly the EW6.40, EW6.68, and EW6.97 andrelative ratio, are well reproduced by the combined model ofthe mCVs, non-mCVs, and ABs (Figure 9). This is consistentwith the point source origin proposed by Revnivtsev et al.(2009), Hong (2012). The major fraction is occupied by non-mCVs, in contrast to Revnivtsev et al. (2009), Hong (2012).The scale heights (SHs) of the Fe I-Kα (SH6.40), Fe XXV-

Heα (SH6.68), and the Fe XXVI-Lyα (SH6.97) are ∼150, ∼300,and ∼300pc, respectively (Yamauchi et al. 2016). These SHsare consistent with those of the mCVs, non-mCVs, and ABs,which also supports the idea that the origin of GBXE is theassembly of point sources, mainly non-mCVs (for Fe XXV-Heα, Fe XXVI-Lyα, and the 5–10 keV band flux), partly mCVs(for Fe I-Kα) and ABs (for Fe XXV-Heα). The residual at∼8.3keV corresponds to Fe XXV-Heγ and/or Fe XXVI-Lyβ,which will be discussed in Section 4.4.

4.3. Galactic Ridge X-Ray Emission (GRXE)

Ebisawa et al. (2005) resolved ∼10% of the GRXE flux intopoint sources above the detection threshold of ∼2×1031

erg s−1(2–10 keV) in the deepest observation of the GRXEfield at * * ~ -l b, 28 .5, 0 .0( ) ( ). The resolved fraction of thesame region by Revnivtsev & Sazonov (2007) is about 20%above the detection threshold of ∼1031 erg s−1(1–7 keV).These differences are hard to be absorbed by the difference ofthe detection threshold, and hence may be regarded as asystematic error in the estimation of the point source fraction.We have obtained a high-quality GRXE spectrum, which

includes the Fe I-Kα, Fe XXV-Heα, Fe XXVI-Lyα, Fe I-Kβ,Ni XXVII-Heα, Fe XXV-Heγ, and Fe XXVI-Lyγ lines. Unlike theGBXE, the GRXE spectrum cannot be well fitted withany combination of mCVs, non-mCVs, and ABs (χ2/dof=282/91). Large excesses are found at the Fe I-Kα andFe XXV-Heα lines (Figure 10(a)). Taking the continuum fluxinto account, we estimated that the Fe I-Kα and Fe XXV-Heαfluxes of the GRXE are ∼2 and 1.2 times larger than that

Figure 5. Averaged spectra of the mCVs, non-mCVs, ABs for ModelA (see Section 3.3), which are normalized as a point source located at 8kpc.

6

The Astrophysical Journal, 833:268 (10pp), 2016 December 20 Nobukawa et al.

強磁場激変星弱磁場激変星コロナ星の連星系

from any combination of the mCVs, non-mCVs, andABs. Therefore, a significant fraction of GCXE is not dueto the mCVs, non-mCVs, and ABs. A possible originwould be the diffuse emission produced by high activitynear the GC, such as the past big flares of Sgr A*.

The authors are grateful to all members of the Suzaku teamwho provided us with the excellent spectral data. This work issupported in part by the Grant-in-Aid for Scientific Research ofthe Ministry of Education, Science, Sports, and Culture (No.15H02090, MN; No. 24540232, SY; No. 24540229, KK).KKN is supported by the Research Fellow of Japan Society forthe Promotion of Science for Young Scientists.

REFERENCES

Byckling, K., Mukai, K., Thorstensen, J. R., & Osborne, J. P. 2010, MNRAS,408, 2298

Capelli, R., Warwick, R. S., Porquet, D., Gillessen, S., & Predehl, P. 2012,A&A, 545, A35

Cunha, K., Sellgren, K., Smith, V. V., et al. 2007, ApJ, 669, 1011Ebisawa, K., Tsujimoto, M., Paizis, A., et al. 2005, ApJ, 635, 214Hong, J. 2012, MNRAS, 427, 1633Inui, T., Koyama, K., Matsumoto, H., & Tsuru, T. G. 2009, PASJ, 61, S241Koyama, K., Awaki, H., Kunieda, H., et al. 1989, Natur, 339, 603Koyama, K., Hyodo, Y., Inui, T., et al. 2007a, PASJ, 59, S245Koyama, K., Inui, T., Hyodo, Y., et al. 2007b, PASJ, 59, S221Koyama, K., Maeda, Y., Sonobe, T., et al. 1996, PASJ, 48, 249Koyama, K., Tsunemi, H., Dotani, T., et al. 2007c, PASJ, 59, S23Koyama, K., Uchiyama, H., Hyodo, Y., et al. 2007d, PASJ, 59, S237Kushino, A., Ishisaki, Y., Morita, U., et al. 2002, PASJ, 54, 327Launhardt, R., Zylka, R., & Mezger, P. G. 2002, A&A, 384, 112Muno, M. P., Baganoff, F. K., Bautz, M. W., et al. 2003, ApJ, 589, 225Muno, M. P., Baganoff, F. K., Bautz, M. W., et al. 2004, ApJ, 613, 326Muno, M. P., Bauer, F. E., Bandyopadhyay, R. M., & Wang, Q. D. 2006,

ApJS, 165, 173Nakajima, H., Tsuru, T. G., Nobukawa, M., et al. 2009, PASJ, 61, S233Nakashima, S., Nobukawa, M., Uchida, H., et al. 2013, ApJ, 773, 20Neustroev, V. V., & Zharikov, S. 2007, MNRAS, 386, 1366

Nobukawa, M., Koyama, K., Tsuru, T. G., Ryu, S. G., & Tatischeff, V. 2010,PASJ, 62, 423

Pandey, J. C., & Singh, K. P. 2012, MNRAS, 419, 1219Park, S., Muno, M. P., Baganoff, F. K., et al. 2004, ApJ, 603, 548Patterson, J. 1984, ApJS, 54, 443Ponti, G., Terrier, R., Goldwurm, A., Belanger, G., & Trap, G. 2010, ApJ,

714, 732Reid, M. J., & Brunthaler, A. 2004, ApJ, 616, 872Revnivtsev, M., & Sazonov, S. 2007, A&A, 471, 159Revnivtsev, M., Sazonov, S., Churazov, E., et al. 2009, Natur, 458, 1142Revnivtsev, M., Vikhlinin, A., & Sazonov, S. 2007, A&A, 473, 857Revnivtsev, M. G., Molkov, S., & Sazonov, S. 2006a, MNRAS, 373, L11Revnivtsev, M. G., Sazonov, S., Gilfanov, M., Churazov, E., & Sunyaev, R.

2006b, A&A, 452, 169Sakano, M., Koyama, K., Murakami, H., Maeda, Y., & Yamauchi, S. 2002,

ApJS, 138, 19Sazonov, S., Revnivtsev, M., Gilfanov, M., Churazov, E., & Sunyaev, R. 2006,

A&A, 450, 117Serlemitsos, P. J., Soong, Y., Chan, K., et al. 2007, PASJ, 59, S9Sheets, H. A., Thorstenaen, J. R., Peters, C. J., Kapusta, A. B., & Taylor, C. J.

2007, PASP, 119, 494Tanuma, S., Yokoyama, T., Kudoh, T., et al. 1999, PASJ, 51, 161Tawa, N., Hayashida, K., Nagai, M., et al. 2008, PASJ, 60, S11Terrier, R., Ponti, G., Bélanger, G., et al. 2010, ApJ, 719, 143Tsuboi, M., Handa, T., & Ukita, N. 1999, ApJS, 120, 1Tsujimoto, M., Hyodo, Y., & Koyama, K. 2007, PASJ, 59, S229Uchiyama, H., Nobukawa, M., Tsuru, T. G., & Koyama, K. 2013, PASJ,

65, 19Uchiyama, H., Nobukawa, M., Tsuru, T. G., Koyama, K., & Matsumoto, H.

2011, PASJ, 63, S903Vaiana, G. S., Cassinelli, J. P., Fabbiano, G., et al. 1981, ApJ, 245, 163Wargelin, B. J., Beiersdorfer, P., Neill, P. A., Olson, R. E., & Scofield, J. H.

2005, ApJ, 634, 687Warwick, R. S. 2014, MNRAS, 445, 66Wienen, M., Wyrowski, F., Menten, K. M., et al. 2015, A&A, 579, 91Worrall, D. M., Marshall, F. E., Boldt, E. A., & Swank, J. H. 1982, ApJ,

255, 111Xu, X., Wang, Q. D., & Li, X. 2016, ApJ, 818, 136Yamauchi, S., Ebisawa, K., Tanaka, Y., et al. 2009, PASJ, 61, S225Yamauchi, S., & Koyama, K. 1993, ApJ, 404, 620Yamauchi, S., Nobukawa, K. K., Nobukawa, M., Uchiyama, H., &

Koyama, K. 2016, PASJ, 68, 59Yasui, K., Nishiyama, S., Yoshikawa, T., et al. 2015, PASJ, 67, 123Yuasa, T., Makishima, K., & Nakazawa, K. 2012, ApJ, 753, 129

Figure 10. (a) GRXE spectrum fitted with the combination of the mCVs (orange), non-mCVs (blue), and ABs (red) of ModelB. The black solid curve shows the CXBmodel. (b) The same as (a), but for the GCXE.

10


リッジ

●点源の足しあわせではスペクトルを説明できない

●少なくとも6.4 keV輝線の半分は別起源

●スケールハイトを考えると、分子ガス起源だろう

混合モデル

X線照射起源？13

X線源の積分光度 6×1038 erg/s (Grimm+02, Molaro+14)

このうち0.005%が6.4 keVに変換

明るいX線源が星間ガスを照らしている？(Molaro+14)

観測値 = 3×1035 erg/s X線照射の寄与は10%

=> X線照射による6.4 keV光度 3×1034 erg/s

分子ガス起源の多くは宇宙線照射

Yamauchi, KN+ 2016

10 kpc

100 pc

NH = 3−4×1022 cm−2

まとめ 14

● 「すざく」で鉄輝線を3本に分離し、銀河面全体を観測 ● 銀河中心東側と銀経-20度付近で、6.4 keV分布の超過を発見。--> 低エネルギー宇宙線起源の可能性が高い ● 銀河面の6.4 keVのスケールハイトは分子ガスと無矛盾 ● 点源のスペクトルの混合では6.4 keV輝線は高々半分しか説明できない。 ● 銀河面の6.4 keVのかなりの割合が低エネルギー宇宙線起源

KN et al. ApJ, 807, L10 (2015) Yamauchi, KN et al. PASJ, 68, 59 (2016) Nobukawa, KN et al. ApJ, 833, 268 (2016)

銀河中心・バルジ・リッジX線放射 15

GCXE region is confirmed by the direct infrared star-countingobservation of the SIRIUS by Yasui et al. (2015).

The EW6.40, EW6.68 and EW6.97 of the GCXE are all largerthan those of the mCVs, non-mCVs, and ABs (Tables 3 and 1).In fact, the fit of the GCXE spectrum by a combination of themCV, non-mCV, and AB spectra is completely rejected withthe large excesses in the Fe I-Kα, Fe XXV-Heα, and Fe XXVI-Lyα lines (Figure 10(b)). The Fe I-Kα, Fe XXV-Heα, andFe XXVI-Lyα fluxes of the GCXE (Table 1) are, respectively,∼2.0, 1.2, and 1.3 times larger than those estimated from thebest-fit combined model (Model B). These excess ratios(relative flux) are larger than possible systematic relativeerrors, and hence the excesses of iron lines are robust results.

The Fe I-Kαline is due to cold gas, while the Fe XXV-Heαand Fe XXVI-Lyα lines are attributable to hot plasma. Thus, asignificant contribution of additional components is requiredregardless of whether diffuse or other point sources. Thiscomponent should emit stronger K-shell lines of iron than anyother known categories, and simultaneously satisfy apparentlyopposite characteristics: excess of cold gas (Fe I-Kα) and thatof hot plasma (Fe XXV-Heα and Fe XXVI-Lyα).

The SH6.40, SH6.68, SH6.97, and that of the 5–10keV bandflux of the GCXE are all similar to ∼30–35pc (Yamauchiet al. 2016). These are far smaller than those of the mCVs, non-mCVs, and ABs, and are more like that of the central molecularzone (CMZ) (Tsuboi et al. 1999; Wienen et al. 2015). There-fore, the origin of the GCXE may be closely related to theCMZ. Near Sgr A* ( * 1l 0 .3∣ ∣ ), the longitude profile ofFe XXV-Heαin the east (positive l*) shows a significant excess

over the west, even if we exclude the bright supernova remnantSgr A East. This excess would be due to larger populations ofhigh-mass stars in the east than the west (Muno et al. 2004;Park et al. 2004; Koyama et al. 2007a). These high-mass starsmay contribute to the GCXE by possible starburst activity andfrequent supernova in the CMZ. Another possibility would bemagnetic reconnection in the CMZ with strong magnetic field,similar to the GRXE (Tanuma et al. 1999).Big outbursts of Sgr A* in the past (Inui et al. 2009; Ponti

et al. 2010; Terrier et al. 2010; Capelli et al. 2012) would makea very hot plasma and LECRs, which may make ionized ironshigher than that in normal plasmas near the GC. Then thetransitions of highly excited level ( >n 2) to the ground state(n= 1) are more enhanced compared to the CIE plasma. Line-like residuals at ∼7.8–7.9 keV and ∼8.2–8.3 keV, would besuch enhanced iron lines. Nakashima et al. (2013) discussed apossible effect of past big flares of Sgr A* on a plasmaspectrum at the south of the GC. The residual at ∼7.6keVfound in the GCXE and GRXE corresponds to Ni I-Kα,because unlike Fe I-Kα and Fe I-Kβ, Ni I-Kα is not included inthe Model B spectra.

5. SUMMARY

The summary for the origin of the GDXE based on the EWsand SHs of iron K-shell lines of the GCXE, GRXE, GBXE,mCVs, non-mCVs, and ABs are as follows:

1. EW of the iron K-shell lines (EW6.40, EW6.68, andEW6.97) and their intensity patterns are different betweenthe GCXE, GRXE, and GBXE.

2. The X-ray spectrum near the iron K-shell lines of theGBXE is well explained by the non-mCVs dominantplasmas with a small contribution of mCVs and ABs.

3. The X-ray spectrum near the iron K-shell lines of theGRXE shows significant excess at the 6.4 keV line(EW6.40) from any combination of the mCVs, non-mCVs,and ABs. The excess of EW6.40 is likely due to low-energy cosmic protons.

4. The X-ray spectrum of the GCXE shows significantexcesses of Fe I-Kα, Fe XXV-Heα, and Fe XXVI-Lyα

Figure 9. (a) GBXE spectrum fitted with the combination of the mCVs (orange), non-mCVs (blue), and ABs (red) of the Model A. In this case, mCVs rarelycontribute to the 5–10keV spectrum. The black solid curve shows the CXB model. (b) The same as the left panel, but fitted with ModelB.

Table 5Best-fit Flux Ratio of mCVs, non-mCVs, and ABs for Model B

GBXE GRXE GCXE

fmCV <0.03 0.09( ) 0.10±0.05 0.04±0.01fnon-mCV 0.67±0.06 0.51±0.06 0.96±0.01fAB 0.30±0.03 0.39±0.02 <0.00 0.01( )

χ2/dof 148/95 (1.56) 282/91 (3.10) 2637/276 (9.55)

9

The Astrophysical Journal, 833:268 (10pp), 2016 December 20 Nobukawa et al.バルジ

強磁場激変星

弱磁場激変星

連星系



REFERENCES


















10


銀河中心

強磁場激変星

弱磁場激変星



REFERENCES


















10


リッジ

強磁場激変星

弱磁場激変星連星系

●銀河中心: 　6.4、6.7、6.9 keVで大きな超過 ●バルジ: 点源で説明できる　(チャンドラ衛星の結果と無矛盾) ●リッジ: 6.4、6.7 keVで超過

銀河中心・リッジは (既知の) 点源だけでは説明不可

M. Nobukawa, KN+2016

銀河リッジ x 線放射の中性鉄輝線 における低エネ...

Documents

銀河リッジ x 線放射の中性鉄輝線における低エネ...